Chapter 9 The verdict: theorem, tools, and remainder¶

Definition requires identity and difference; intrinsic nature cancels identity and difference; hence contradiction. After eight chapters, this sentence can stand at the head of the essay without being a preview. For those who speak and assert, it has a twin: assertion requires identity and difference; intrinsic nature cancels identity and difference; and hence the assertion defeats itself. This chapter convenes the court in three sections. The first accepts the theorem: the premise audit and the exchanges over objections are laid out before the court, and the verdict is settled. The second hands over the tools: pending cases recorded in the preceding three essays are handled one by one under the theorem. The third inventories the remainder: once the theorem and tools are extracted from the Mūlamadhyamakakārikā, what remains? This book gives its valuation.

Section 1: accepting the theorem. Acceptance begins with the premises. The proof has used three classes of premise, inventoried item by item. Appendix A is the master record, with the full table of sources and uses. First, four definitions: D1, the three senses of existence; D2, the three conditions for intrinsic nature; D3, the scope of application; and the clause delimiting definition. Definitions answer for fidelity, not truth or falsity. Chapter 2 delivered that fidelity word by word. All three criteria came from texts the opponents themselves used. What is refuted is no straw man constructed by this essay. Second, one body of logic: classical logic, together with the clauses limiting the use of the law of noncontradiction and Leibniz's law. Chapter 3 checked their use, with no departure from the classical throughout. Third, one analytic truth: P, that whatever exists necessarily has identity conditions involving difference and relations. The three-step argument, the specification of granularity, the grading of "dependence," and the four closing routes are all in Chapter 4's record. There is no fourth class.

One absence from the inventory must be explicit: not a single instance of scriptural authority. The original texts have three duties in this essay, none at the place of premises: supplying sources for definitions, historical instances for reductio, and evidence from forms of expression for readings. Remove all the Buddhist material and the theorem still stands. This is no declaration. Chapter 8 demonstrated it on the spot.

The objections and replies are filed in one table. Chapter 7's four modes of failure form the main record, together with three questions arriving from the side. Each row records the charge, the location of the reply, and the finding.

The verdict. Exit one holds: T-strong holds at analytic strength, grounded in the dependence axis both parties acknowledge and the domain of conditions. Record the two versions separately. Those who state a determinate nature collide with contrastive dependence at the type level and belong to the strong version. Those who retreat to the unstatable are taken over by the apparatus at the level of acts and belong to the middle version. T-weak, as programme and established results, does not depend on the fate of the first two versions. Exits two and three have not been triggered. No one has yet demonstrated noncontrastive individuation; its very description already uses contrast. Every channel for downgrading stays open. This is the honest form of the essay's statement of its force. Chapter 2 set out the costs of conditional downgrades and of taking the way out offered by overturning the table. Whoever rejects a premise receives the corresponding grade of conclusion. All three exits yield results. The final sentence of the verdict confirms the outcome recorded in Chapter 7: we looked seriously for gaps, found one, and repaired it on the spot; we also gained three further clarifications—the theorem's stated domain, the invariant across frameworks, and the relational character of independence itself.

Section 2: handing over the tools. Essay 15's final chapter drew up an inventory expressly left for the following essays to use. Two items were assigned to Madhyamaka: intrinsic nature, and the criterion that posits something as a real dharma if it cannot be omitted from a sentence. This section receives them against that inventory, then joins them to the pending cases in the verdict chapters of Essays 16 and 17. The procedure is uniform: first give the D3 form, then identify the tool used, and finally record what is preserved at the ordinary level.

First, the entity of a dharma really existent in the three times. D3 form: there is a dharma entity that survives decomposition and analysis, exists permanently, and does not change with conditions. Essay 15, Chapters 3 and 7 recorded all its difficulties. The differences among the three times depend entirely on function; function has no identity; the entity and its function cannot be determinately said to be identical or different. The theorem gathers this knot of difficulties into one step. Differences among the three times are properties of contrastive dependence. Unity and difference arise through mutual dependence, as the "Examination of Time" says, whereas the dharma entity's pledge is precisely independence. Function's lack of identity is the formal manifestation of the clash between pledge and duty. It falls exactly at the mark fixed by the scope of application and is no Sarvāstivāda oversight. On the preservation side, the whole level of discourse about the three times remains unharmed. Time continues to work as an E2 network of mutual dependence. Essay 15 said that the end of the chain supported a word whose name had changed six times. This chapter adds the verdict: no name can be found for that place, because the theorem makes any occupant of it impossible.

Second, acquisition and nonmanifest material form, together with the life faculty and the commonality of sentient beings. Essay 15 read the grounds for establishing this family as a criterion: if something cannot be omitted from a sentence that must hold, posit its real existence, at a one-to-one exchange rate. The theorem handles it by exactly the boundary drawn in Chapter 7. There is no such inference from the level of conditions to the level of existence. Indispensability proves the need for registration at E2, not the existence of an E3 foundation. The regress of acquisition-of-acquisition then stops. Once an entry need not really exist, there is no reason to register the entry again. The regress relied precisely on the requirement that every link be a real dharma. On the preservation side, discourse about possession, keeping precepts, and being of the same kind continues. The two Chinese characters 假立, "nominal positing," on Essay 17's verdict sheet now receive a warrant in the form of a theorem.

Third, atoms. Essay 15, Chapter 6 delivered the cleanest example: all four clauses are determinations, and each added determination requires another to fill the gap it leaves. D3 form: there is a smallest material dharma that survives decomposition and analysis and depends on no other dharma. Disposition: the atom's whole identity is supplied by such determinations of contrastive dependence as not touching, adjacency without gaps, and the co-arising of eight items. Its identity conditions are a network of pure contrastive dependence. Pledge and network collide, entering exactly the third and fourth steps of the core derivation. On the preservation side, discourse about granularity remains unharmed. Minimal units are legitimate items at E2, and they never pledge independence.

Fourth, the pudgala. The Vātsīputrīya maintain that the self is neither identical with nor different from the aggregates and is therefore inexpressible. Both routes have been judged. If inexpressibility is a refusal to state anything, it leaves the thesis domain and is taken over by the apparatus at the level of acts. If inexpressibility is a determination of the self, "neither identical nor different" remains a predication. π is untouched, and all four statements are rejected as before. Chapter 3's test against the scale and the judgement in the "Refutation of the Self" in Essay 17, Chapter 6 converge here. Essay 15's final chapter recorded what the no-self argument preserved. This chapter adds only the final step in theorem form.

Fifth, nirvāṇa as an entity. Essay 15's final entry recorded the difficulty of the three unconditioned dharmas. Outside the three times, the criterion of function cannot be applied at all, leaving only half the warrant for real existence. Essay 17 recorded the three schools' deadlock and announced that it would have to wait for Madhyamaka to address the categories of existence and nonexistence themselves. The "Examination of Nirvāṇa" fulfils the announcement. Existence, nonexistence, both, neither: all four routes are rejected, ending with "This cannot be so." The categories themselves are handled in the "Examination of Existence and Nonexistence," from which this essay's Chapter 1 takes its definitions. Nirvāṇa is no E3-plus-D2 object. If it were, part (b) of Chapter 6's preservation theorem would make it wholly inert, unable to enter any causation involved in practice: "If there were no emptiness, the unattained could not be attained."¹ On the preservation side, discourse about the path and its fruits remains unharmed. Practice and liberation are among what is preserved, as Chapter 6 proved. This pending case is not answered but withdrawn. Asking whether nirvāṇa exists or does not exist asks in the wrong sense of existence.

Sixth, bringing the cases together. Essay 16's final chapter judged that commentary becomes a system through rulings because argument cannot achieve it. This chapter can complete the causal account. Argument cannot do it because the task is logically impossible, not because the treatise masters lack ability. What the rulings protect is a D3-type combination. Every attempt to establish it by argument brings this inference into play. The verdict chapters of the three essays now join in one line. The Mahāvibhāṣā must fit itself with brakes, in Essay 16. The Abhidharmakośabhāṣya replaces the brakes with a steering wheel, in Essay 17. This essay states that the vehicle has nowhere to go: every route towards E3 plus D2 ends in contradiction. Yet the tools aboard, classification, causes and conditions, and continuity, can travel every road at the level of E2. This is precisely what handing them over means: no confiscation, just new registration plates.

Section 3: an inventory of the remainder. Take the theorem and tools out of the Mūlamadhyamakakārikā. What remains? The question matters because, if the answer is nothing, this treatise is only a book devoted to reductio. Reading it would be no different from reading an exercise book. This book's answer has three levels, each valued more highly than the last.

The first level is positive teaching. The homage verse writes the purpose of the whole treatise into its closing line: "Able to teach these causes and conditions, skilfully bringing all conceptual proliferation to an end."² Dependent origination is what is preserved, not what is proved. None of the twenty-seven chapters tries to prove dependent origination. They prove that there is nowhere to stand outside it. The first remainder is therefore the positive teaching of the whole E2 level: the sequence of twelve links of dependent origination, with the entire "Examination of the Twelve Links of Dependent Origination" consisting of positive teaching and not a single word of refutation; the continuity of karman and its results; the stages of path and fruit. The theorem removes the demand for a foundation. It does not supply one.

The second level is the direction of practice. The "Examination of Dharmas" gives an entry: "In the true character of dharmas, mental activity and speech cease." Qingmu's commentary in the same chapter says, "Since there is no dharma here to grasp or discard, it is called the character of quiescent cessation." It also records that those dwelling in the true character "are not made the playthings of conceptual proliferation."³ "True character" names a state in which conceptual proliferation ceases, not a deeper object. In the theory of practice, this means one thing. The theorem's completed form is a state in which holding on is no longer needed. It does not stop at a conclusion held on to. Proof is a road, not a residence.

The third level is the final step. Chapter 7 judged that the theorem fulfils its duty by remaining silent in the dimension of situations. A meta-constraint's job is to issue rules, not describe situations. Now add this step: the theorem's acknowledgement of its own silence is the semantic form of "emptiness too is empty." Emptiness is a rule, not information about the world. The rule says everything it should within every framework and hands in a blank sheet for the level of situations. That blank sheet is precisely where its duty lies. Two classical statements now take their places without impinging on each other. "Mental activity and speech cease" concerns the level of situations. No meta-constraint had a speaking part there in the first place. Speech ceases there exactly where it should. "The ultimate always depends on speech. Speech is conventional"⁴ concerns the meta level. Rules are written in conventional discourse. One statement leaves words behind, the other relies on them. They belong to two different dimensions, without contradiction anywhere.

Close with the oldest analogy, the raft.⁵ Crossing the river requires a raft. Having reached the shore, one does not carry the raft onward. The theorem is a raft. All its efficacy lies in the crossing: reject every D3 claim, preserve E2. It is not itself furniture for the other shore. Whoever walks with the raft on their head will find level clause (c) waiting on shore. This book's valuation of the remainder is now fixed. Once the theorem and tools are taken away, the Mūlamadhyamakakārikā retains what it truly meant to give from beginning to end: a path needing no metaphysical foundation. This conclusion agrees with the conclusion of modern research. Westerhoff's final chapter likewise reads Nāgārjuna as offering a positive philosophical project, not a bound collection of reductios.⁶

Passing the baton. In India, this proof was read by two schools. One held that reductio is its entire syntax and that no inference may be established beyond negation. The other held that emptiness must be positively established by autonomous inference, positing after negation. Their dispute can be stated in one sentence with this essay's notation: can the unified authorization ⊢¬π be obtained only through reductio, or can it be established directly? China did not wait for this dispute to arrive before opening its own arena over the two truths. Are the two truths principles or teachings, one or different, and how are their successive levels established? The next essay opens from these two points. This essay closes by handing over three inventories: the premises in Appendix A, the tools in Section 2, and the remainder in Section 3.

A restatement in plain language follows the same rule as before: change only the language, add no content, and let the first half govern any discrepancy.

Definition requires identity and difference; intrinsic nature cancels identity and difference; hence contradiction. After eight chapters, this no longer needs to be a preview. It can stand right at the opening. For those who speak, there is a twin version: assertion requires identity and difference; intrinsic nature cancels identity and difference; and hence the assertion defeats itself. This chapter convenes the court in three parts. First accept the theorem, put every premise used and every challenge faced on the table, then deliver the verdict. Next hand over the tools and settle the old cases left pending by the preceding three essays under the theorem, one by one. Finally count our possessions. Once the theorem and tools are moved out, what does the Mūlamadhyamakakārikā have left?

First, acceptance. Our proof has used only three classes of premise. The first is definitions, four of them: the three senses of existence, the three conditions for intrinsic nature, the scope of application, and a boundary clause saying where definition stops. Definitions answer for honesty, not truth or falsity. Chapter 2 checked their honesty word by word against the original texts. All three criteria were used by the opponents themselves. We did not construct a straw man. The second class is logic, a whole system of classical logic, together with the clauses limiting how the law of noncontradiction and Leibniz's law are used. Chapter 3 checked it, and it has stayed on course. The third class has only one member, that truth of conceptual grammar: making clear how a thing is itself cannot avoid other things. Chapter 4 has the full case on why it holds and at what strength. Just these three classes, no fourth.

What is missing from the inventory matters still more. No saying of the Buddha serves as a premise. Sūtras and treatises do only three jobs here: sources for definitions, historical material for reductio, and evidence from forms of expression for readings. Remove all the Buddhist material and the theorem still stands. This is not just talk. Chapter 8 demonstrated it on the spot.

The challenges faced number seven. List them. Surely, in the end, there is a "this one" for which no reason can be stated? Reply: Primitive does not mean inexpressible. If a primitive item is assigned the task of distinguishing A from B, it falls back into relational dependence; bare thisness fails the E3 criteria. If an opponent insists that identity facts come first and facts of difference follow, and says no more, the E3 route and the middle version take over. This remaining case is recorded under the middle version, not counted again under the strongest version. Have you given relations the place of entities? Reply: a solidly real relation cannot reach both ends. Relations are all at the ordinary level, and this challenge yields another conclusion. If identity conditions all involve relations, does nothing have a chance of solidly real existence? Reply: we posit no inference from identity conditions to existence. Chapter 7 drew the boundary. Twenty-seven cases cannot yield an induction about everything? Reply: we do not count cases; the force comes from conceptual grammar. If your theorem is true in every framework, has it said nothing? Reply: there are two kinds of having nothing to say. What cannot be thrown away is a rule; only what can be thrown away is empty talk. Involving other things and depending on them are different, surely? Reply: the dependence rejected by the pledge of intrinsic nature includes contrastive dependence in both sides' original texts. Both sides already share this point. The nominal must rest on the real, surely? Reply: propping things up is work, and what does no work cannot provide a foundation. All seven challenges have been met, and none has forced a downgrade.

So the verdict is that the first exit holds. The strongest version holds at analytic strength, grounded in the axis of dependence acknowledged by both parties. Record the two versions separately. Whoever states a determinate nature collides and is entered under the strongest version. Whoever stubbornly refuses to speak is taken over by the apparatus at the level of acts and entered under the middle version. The weakest version's results were already there and do not live off the first two. The second and third exits have not been triggered. To date no one has demonstrated a way of identifying things without comparison. The sentences describing it themselves compare. Of course, the doors to downgrading stay open. This is the honest way to state our argument's force. Chapter 2 priced every retreat. Reject a premise and receive the corresponding grade of conclusion. Every one of the three exits yields something. One last sentence confirms Chapter 7's verdict: we looked seriously for gaps, found one, and fixed it on the spot; we also came away with three things made clearer.

Next, hand over the tools. Essay 15 ended with an inventory promised for the later essays to use. The two items assigned to Madhyamaka were intrinsic nature and the criterion that counts whatever cannot be omitted as really existent. We now receive them against the inventory and handle them one by one, together with the old cases pending in the verdict chapters of Essays 16 and 17.

Dharma entities. Essay 15 recorded all the difficulties of the entity really existent in the three times. The differences among the three times depend entirely on function, yet function has no identity. The theorem gathers it into one step: differences among the three times are properties that arise through mutual dependence, whereas the entity's pledge is precisely to depend on nothing else. Pledge and duty conflict exactly at the point caught by fixing the scope of application. On the ordinary side, all discourse about time continues, with nothing lost. Essay 15 said that the explanatory chain rests at its end on a word whose name has changed six times. We can now add the verdict: no name can ever be found for that place, since the theorem makes it impossible for anything to occupy it at all.

Acquisition, nonmanifest material form, the life faculty, and the commonality of sentient beings. This family's ground for being established is that whatever cannot be omitted from a sentence is posited as really existent, a one-to-one exchange. The theorem replies with that boundary: indispensability does not entail real existence. It proves only that the ordinary level needs this entry. An entry need not be solidly real, so there is no reason to register the entry again. The infinite chain of acquisition-of-acquisition breaks on its own. Possession, keeping precepts, being of the same kind: all this discourse continues. The two Chinese characters 假立, "nominal positing," in Essay 17's verdict now receive the theorem's warrant.

Atoms. Essay 15, Chapter 6 offered the cleanest example: all four clauses are determinations, and each added determination calls for another. The theorem's disposition: the atom's entire identity is a network of determinations, every mesh mutually dependent, while the pledge insists on depending on nothing else. It collides exactly with steps three and four of the main argument. Discourse about minimal units continues. They work at the ordinary level and have never made that pledge.

The pudgala. The Vātsīputrīya say that the self and the five aggregates are neither identical nor different, hence inexpressible. Both routes are blocked. Truly say nothing, and one leaves the thesis domain; once one speaks, the apparatus at the level of acts takes charge. If "neither identical nor different" is taken as a determination of the self, it is still predication. The underlying presupposition has not moved, and all four statements are refuted as before. Chapter 3's test against the scale and the judgement in the "Refutation of the Self" in Essay 17, Chapter 6 converge here.

Nirvāṇa. Essay 15's final entry recorded the embarrassment of the three unconditioned dharmas. Outside the three times, the criterion of function is wholly inapplicable, leaving half the warrant for real existence missing. Essay 17 said that this deadlock must wait for Madhyamaka to handle the categories of existence and nonexistence themselves. The "Examination of Nirvāṇa" fulfils the promise: existence, nonexistence, both, neither, all four routes refuted, ending with "This cannot be so." The categories themselves are handled in the "Examination of Existence and Nonexistence," the source of our Chapter 1's definitions. Nirvāṇa is no thing solidly real and established on its own. If it were, the other half of the preservation theorem would make it utterly inert, unable to enter any cause and effect involved in practice. The text says that without emptiness, what should be attained cannot be attained. On the ordinary side, not one statement about cultivating the path and attaining its fruits is lost. Chapter 6 preserved them. This pending case is not answered but withdrawn. Asking whether nirvāṇa exists or does not exist asks in the wrong place.

Bring it together in one statement. Essay 16's verdict was that commentary becomes a system through rulings because argument cannot do it. The reason can now be added. When argument cannot do it, do not blame the treatise masters' abilities. The task is logically impossible. The verdicts of three essays join in one line: the Mahāvibhāṣā must bring its own brakes; the Abhidharmakośabhāṣya replaces the brakes with a steering wheel; this essay states that the vehicle has nowhere at all to go. Drive towards solid reality plus being established on its own and every destination is contradiction. Yet the tools aboard, classification, causes and conditions, continuity, can travel every road at the ordinary level. This is what the handover transfers: the tools are not confiscated but given new registration plates.

Finally, count the possessions. Take the theorem and tools out of the Mūlamadhyamakakārikā. What is left? If nothing, the treatise is an exercise book. The answer has three levels, each worth more than the last. First, positive teaching. The opening homage verse states the aim clearly: "Able to teach these causes and conditions, skilfully bringing all conceptual proliferation to an end." Dependent origination is preserved throughout and proved nowhere. Not one of the twenty-seven chapters proves it. They prove that there is nowhere to stand outside it. So the first remainder is the positive teaching of the whole ordinary level: the sequence of twelve links of dependent origination, the continuity of karman and its results, and the steps of cultivating the path. The theorem does not lay a foundation for them. It removes the demand for a foundation itself.

Second, the direction of practice. The "Examination of Dharmas" says that in the true character of dharmas, mental activity and speech cease. The commentator adds that with nothing to grasp or discard, this is called the character of quiescent cessation. "True character" lies no level deeper. It names the state in which conceptual proliferation stops. In practice, this means that the theorem's completed form is a state in which there is no longer any need to grip anything. Whether one grips that conclusion no longer matters. Proof is a road, not a place to live.

Third, the last step. Chapter 7 judged that the theorem fulfils its duty by having nothing to say at the level of situations. A rule's job is to issue rules, not describe situations. Now add this step: the theorem acknowledges its own silence. This is what "emptiness too is empty" says. Emptiness is a rule, not a piece of information about the world. The rule says everything it should within each framework and hands in a blank sheet for the level of situations. Duty requires that blank sheet; it is no shortfall. Two old statements take their respective places here. "Mental activity and speech cease" concerns the level of situations, where rules never had a speaking part. "The ultimate always depends on speech" concerns the level of rules, which are written in conventional language. One says to leave words behind; the other says to rely on them. Two levels apart, they do not conflict.

End with that oldest analogy, the raft. Crossing a river needs a raft. Once ashore, one does not carry the raft. The theorem is a raft. All its ability lies in the crossing: refute every claim of solid reality established on its own and preserve the whole ordinary level. It is not itself furniture for the other shore. Whoever reaches shore and still carries the raft on their head will find Chapter 5's third rule waiting at the water's edge. So our valuation of the remainder is this: take the theorem and tools away, and the Mūlamadhyamakakārikā retains precisely what it truly meant to give from beginning to end, a path needing no metaphysical foundation. This also agrees with where contemporary research has arrived. Westerhoff's final chapter likewise reads Nāgārjuna as offering a positive philosophical project, not a bound heap of reductios.

Pass the baton. In India, the proof divided into two schools of reading. One said that negation is its entire syntax, with no separate form of argument permitted beyond reductio. The other said that emptiness must be positively established through its own form of argument. In our notation, the dispute is whether that unified authorization can be obtained only through reductio or can be proved directly. China did not wait for this dispute to arrive before opening its own arena over the two truths. Are the two truths principles or teachings? Are they one or different? How are the successive steps built? The next essay opens from these two points. This essay closes by handing over three inventories: the premises in Appendix A, the tools in Section 2, and the remainder in Section 3.

References¶